Invariant natural killer T-cells (NKT) are a subset of T-cells that are implicated in a broad range of diseases. In some circumstances they can enhance the response to infection (Kinjo, Illarionov et al. 2011) and cancer (Wu, Lin et al. 2011) but also possess the ability to suppress autoimmune disease (Hong, Wilson et al. 2001) and type II diabetes. Activation of NKT cells can also lead to undesirable immune responses as related to allergy, (Wingender, Rogers et al. 2011) autoimmunity (Zeng, Liu et al. 2003) and atherosclerosis (Tupin, Nicoletti et al. 2004).
Unlike conventional T-cells that are restricted by major histocompatibility complex (MHC) molecules that present peptide antigens, NKT cells are uniquely restricted by CD1d proteins (Bendelac, Savage et al. 2007). CD1d proteins belong to the CD1 family that contains five members, CD1a-e. Like MHC molecules, the CD1 family members all contain an antigen binding region that is flanked by two anti-parallel α-helices that sit above a β-sheet. Unlike MHC molecules, the binding region of the CD1 proteins contain two large hydrophobic binding pockets that are suited to bind lipid antigens rather than peptide-based antigens (Li, Girardi et al. 2010). α-Galactosylceramide (α-GalCer) is the most studied NKT cell antigen and potently activates human and mouse NKT cells (Kawano, Cui et al. 1997). In animal studies, α-GalCer is reported to be useful in the treatment of a number of diseases including cancer, (Morita, Motoki et al. 1995; Motoki, Morita et al. 1995) and autoimmune disease (Hong, Wilson et al. 2001). The compound has also been shown to function as a potent vaccine adjuvant in the treatment and prophylaxis of cancer and infectious disease (Silk, Hermans et al. 2004). This adjuvant activity has been attributed to stimulatory interactions between activated NKT cells and dendritic cells (DCs), the most potent antigen-presenting cells in the body. As a consequence, the DCs are rendered capable of promoting strong adaptive immune responses (Fujii, Shimizu et al. 2003; Hermans, Silk et al. 2003).
There is considerable interest in therapeutic vaccines for the treatment of cancer. The aim is to stimulate clonal expansion of T cells within a host that are capable of recognising and killing tumour cells, leaving normal tissues intact. This specificity relies on recognition of unique, tumour-derived, protein fragments presented by major histocompatibility complex (MHC) molecules on the tumour cell surface. Vaccines used in this context typically involve injection of the defined tumour-associated “tumour antigens”, or their peptide fragments, together with immune adjuvants capable of driving an immune response. In the absence of such adjuvants, the opposite outcome may ensue, with the tumour antigens actually being “tolerated” by the immune system rather than provoking tumour rejection. Advances in this therapy are therefore dependent on appropriate combinations of antigen and adjuvant (Speiser and Romero 2010).
When incorporated into a vaccine, α-GalCer must first be acquired by antigen-presenting cells in the host, and then presented to NKT cells within the local environment (Fujii, Shimizu et al. 2003; Hermans, Silk et al. 2003). This process brings the two cell-types into close association, permitting stimulatory signals to be passed from NKT cell to antigen-presenting cell.

Importantly, if the same antigen-presenting cells acquire the defined antigens of the vaccine, the stimulatory signals received through interaction with NKT cells can be translated directly into a superior capacity to provoke clonal proliferation of antigen-specific T cells with capacity to kill (Hermans, Silk et al. 2003; Semmling, Lukacs-Kornek et al. 2010). One way to achieve this is to load antigen-presenting cells ex vivo with antigenic material and NKT cell ligands (Petersen, Sika-Paotonu et al. 2010). Although a promising approach, in the clinic this requires leukapheresis and the ex vivo culturing of peripheral blood mononuclear cells (PBMC) over 7 days in a highly controlled sterile facility to generate sufficient antigen-presenting cells, which is a cumbersome and costly process. An alternative is to target antigen-presenting cells in vivo, with covalent attachment of antigen to NKT cell ligand ensuring entry into the same cell. Although used successfully with other immune adjuvant compounds, including the covalent attachment of a TLR2 agonist to MUC1 peptides (Cai, Huang et al. 2011), the approach has not been regarded as easily applicable to α-GalCer because the chemical attachment of peptide will result in a conjugate with significantly diminished, or no, capacity to stimulate NKT cells. In particular, the specific lipid moieties of α-GalCer are required for optimal binding into the A and F pockets of CD1d, and the polar head-group is required to be positioned appropriately for interaction with the T-cell receptor of the NKT cell (Borg, Wun et al. 2007), placing particularly tight constraints on the whole glycolipid structure for activity.
Although α-GalCer has considerable biological activity it does have limitations such as poor solubility, (Ebensen, Link et al. 2007) lack of efficacy in human clinical trials, (Giaccone, Punt et al. 2002) promotion of T-cell anergy (Parekh, Wilson et al. 2005) and the generation of both Th1 and Th2 cytokines that may contribute to mixed results in model studies.
It is an object of the invention to provide novel compounds or vaccines useful as agents for treating diseases or conditions relating to cancer, infection, autoimmune disease, atopic disorders or cancer, or to at least provide a useful alternative.